Cognitive impairment and dementia preclude healthy aging. Several lines of evidence suggest that cardiovascular disease (CVD) and its risk factors (and related biomarkers) accelerate decline in cognitive function, but most studies are cross-sectional or have a <5 year follow-up. We propose to use >20 years of CVD and cognitive function test (CFT) data from the Rancho Bernardo Study (RBS) of older community-dwelling adults to examine the association of baseline and changing CVD risk factors with change in CFT performance. The RBS was established as a Lipid Research Prevalence Study in 1972, when history of heart disease, diabetes, stroke, and hypertension was obtained, and body mass index, fasting plasma glucose, triglycerides and cholesterol were measured in 82% of adult residents. Beginning in 1984, 60-87% of surviving community-dwelling participants returned to our research clinic every four years, when CVD risk factors were measured, and lifestyle, medication, and medical history were updated. Beginning in 1988, 3 or more standard CFTs were completed at each of six clinic visits about 4 years apart. To date, 1172 men and women have been retested on at least three occasions, and 721 have had four or more CFT visits. We propose a prospective study of the association of CVD risk factors (4 risk factor groups and 3 composite risk factor scores) and clinical and subclinical CVD with successful cognitive aging. We define successful cognitive aging as survival without clinical dementia and maintenance of predefined performance on CFTs (3 standard tests assessing 3 cognitive domains). We will first establish whether favorable CVD risk factors are prospectively related to short term (4 year) maintained cognitive function, followed by longer term (20 year) trajectory studies which will include change in risk factors as well as change in CFTs. We will examine predictors of categorically-defined successful cognitive aging using Cox proportional hazard models. Associations will be examined in the context of education, depressed mood, weight and weight change, Apo E genotype, medications and comorbidities;mechanisms will be studied using measured biomarkers including CRP, IL-6, IGF-1, steroid hormones, and adipocytokines. All analyses will be based on specific a priori hypotheses. Our goal is to define a plausible web of causation for CVD risk factors and biomarkers as exposure variables, with more favorable attributes predicting less or no loss of cognitive function. Longitudinal studies are the strongest method for addressing the strength, independence, temporality, and effect modification of multiple observed associations. The rich RBS dataset, unique for its long-term repeated CVD &CFT measures, will allow us to determine how CVD exposure variables influence different domains of cognitive function. This information is important given the aging of the population, the fact that CVD risk factors are modifiable, and because the maintenance of cognitive function is an essential component of healthy aging.